A repeat longitudinal, intervention-evaluation study was conducted in 1999 (baseline), 2001 and 2002 in an informal settlement in Soweto, where an improved housing project (relocation to a new housing estate) was implemented in 2000. The aims of the study were to ascertain group and time effects on satisfaction with the personal and environmental domains of quality of life, and determine personal and environmental predictors of life and neighbourhood satisfaction. The sample comprised four groups of residents: relocated (n=26), awaiting relocation (n=58), site tenure allocated (n=154) and a group from the squatter camp (n=133). Residents were asked to rate their satisfaction with the personal domain (e.g., self, family, friends) and the environmental domain (e.g., housing, schools, health services). They also rated their overall life and neighbourhood satisfaction. The group from the squatter camp had the lowest levels of satisfaction with their personal and environmental quality of life. Satisfaction with personal and environmental quality of life decreased in 2001 and increased in 2002. Satisfaction with one's health explained 46% of the variance in life satisfaction in 1999, 39% in 2001 and 24% in 2002. Satisfaction with housing explained 4% of the variance in neighbourhood satisfaction in 1999, 28% in 2001 and 24% in 2002. It was concluded that the group from the squatter camp were the most disadvantaged in their personal and environmental quality of life, good health is essential for life satisfaction and housing is the most important aspect of neighbourhood satisfaction.